Abstract: [Objectives] Driven by the “dual carbon” goals, this study aims to enhance the economic efficiency and energy performance of the cold-end system in a gas-steam combined cycle peak-shaving unit. Taking the M701J unit of a power plant as the research object, a cold-end optimization model based on net power maximization is established. [Methods] By integrating data-driven modeling, thermodynamic theoretical models, and corrected variable-condition experimental data, a dynamic coupling model is developed to characterize the condenser back pressure, turbine incremental power, and circulating water system power consumption. A multi-constraint collaborative optimization framework based on the sequential quadratic programming (SQP) algorithm is proposed. [Results] The nonlinear equilibrium relationship exists between condenser back pressure and circulating water flow rate. By optimizing the circulating water flow, back pressure, and cooling tower fan operation configuration, the net power output can be significantly improved. Validation under seven typical operating conditions shows that the optimized system achieves a net power increase of 140.48-720.92 kW, demonstrating the energy-saving potential of multi-equipment coordinated optimization. [Conclusions] This study provides an actionable approach for improving cold-end system efficiency through quantitative regulation of control boundaries, offering a reference for energy-saving, consumption reduction, and intelligent operation of similar units.

Key words: dual carbon, power system, power plant, gas turbine, combined cycle unit, peak-shaving unit, cold end optimization, energy saving